GLMNet: Graph learning-matching convolutional networks for feature matching

Abstract

Recently, graph convolutional networks (GCNs) have been employed for graph matching problem. It can integrate graph node feature embedding, node-wise affinity learning and matching optimization together in a unified end-to-end model. However, first, the matching graphs feeding to existing graph matching networks are generally fixed and independent of graph matching task, which thus are not guaranteed to be optimal for the graph matching task. Second, existing methods generally employ smoothing-based graph convolution to generate graph node embeddings, in which extensive smoothing convolution operation may dilute the desired discriminatory information of graph nodes. To overcome these issues, we propose a novel Graph Learning-Matching Network (GLMNet) for graph matching problem. GLMNet has three main aspects. (1) It integrates graph learning into graph matching which thus adaptively learns a pair of optimal graphs for graph matching task. (2) It further employs a Laplacian sharpening graph convolution to generate more discriminative node embeddings for graph matching. (3) A new constraint regularized loss is designed for GLMNet training which can encode the desired one-to-one matching constraints in matching optimization. Experiments demonstrate the effectiveness of GLMNet.